Comparing mailing list traffic for R, SAS and S-PLUS

One of the perennial questions about R is "how many users are there?" -- a notoriously difficult question to ask of open-source software, where no records are kept of who downloads or installs it. But on the r-help mailing list recently, Marc Schwartz has come up with an analysis that at least sheds some light on the rate of growth of R usage, if you consider traffic on a discussion mailing list as a proxy for usage.  

Marc analyzed the r-help mailing list and counted the number of distinct posts in each year since the mailing list was created.  The methodology used to collect these data (and those of the charts below) is detailed in Marc's r-help post, but the results expressed as a bar chart are shown here:

Clearly, the rate of discussions on r-help has been increasing consistently every year since it was introduced, and now tops over 30,000 messages per year.

It's interesting to compare this growth with the traffic on the main mailing lists for two major commercial software pages, SAS and S-PLUS.  The traffic on the SAS-L mailing list looks like this:

Traffic has been growing over time, but has been essentially flat for the past 3 years. r-help traffic exceeded SAS-L traffic for the first time in 2008.

S-PLUS is primarily discussed on the s-news mailing list, whose traffic is charted here:

The data and the code Marc wrote to create these charts (modified slightly by me to produce PNGs instead of PDFs) is available: Download mailinglists.R (4.6K).

Update 15:55PST: Thanks to Hadley Wickham, here's all three datasets presented as a time series on the same log-scale using ggplot2:

(Update Jan08 10:10PST: Updated chart above to exclude incomplete 2009 data, from a suggestion by Marc Schwartz.  R code: Download mailinglistcombined.R (4.5K))

Comments

[peng]

hi friends,
I am new to R.I would like to know R-PLUS.Does any know where can I get the free training for R-PLUS.

Regards,
Peng.

[W]

This is very interesting information! Thanks!!

Is it possible to add more data on other software such as MATLAB?